...I make no reference to the attenuation exact value because I'm not sure exactly on it...but should anywhere -20 dB
This is nothing more than the inner conductor of a coax cable passing by the center of an 43 type core with 10 turns giving the tap'ed output.
Signal for the tap attenuator input was provided by the MFJ but had to reduce by -10db the output so it didn't went out of the DVM scale.
Here's the input signal value:
So the math is simple:
Direct signal from the MFJ to a load: 1866mV
Tap'ed value was: 1395mV
So that's: 471mv drop, at 20mV/dB = -23.55dB of attenuation... only the small patch cable attenuation was not accounted... and on my not completely accurate power meter.
Test was only made at around 170Mhz.
There's also the directivity factor that I didn't measured.
Here's one unit similar built by Alan Yates, he got -20db with one less turn (10) than I did (11) and a bigger toroid.
Anyhow in the future I will make another tap attenuator using only resistors.